data <- read_csv("data/data.csv")
##
## ── Column specification ────────────────────────────────────────────────────────
## cols(
## acousticness = col_double(),
## artists = col_character(),
## danceability = col_double(),
## duration_ms = col_double(),
## energy = col_double(),
## explicit = col_double(),
## id = col_character(),
## instrumentalness = col_double(),
## key = col_double(),
## liveness = col_double(),
## loudness = col_double(),
## mode = col_double(),
## name = col_character(),
## popularity = col_double(),
## release_date = col_character(),
## speechiness = col_double(),
## tempo = col_double(),
## valence = col_double(),
## year = col_double()
## )
data <- data %>% mutate(artists= str_split(str_remove_all(artists, regex("\\[|\\]|'")),","))
data <- data %>% mutate(first_artist=map_chr(artists, ~ .[1]))
data %>% group_by(first_artist) %>% summarise(popularity=sum(popularity)) %>% slice_max(order_by=popularity, n=20) %>%
ggplot() +
geom_col(aes(x=reorder(first_artist, -popularity), y=popularity))+
theme(axis.text.x = element_text(angle = 45,hjust = 1)) +
xlab("Artist") +
ggtitle("Artists by total popularity")
data %>%filter(year>=2000) %>% group_by(first_artist) %>% summarise(popularity=sum(popularity)) %>% slice_max(order_by=popularity, n=20) %>%
ggplot() + geom_col(aes(x=reorder(first_artist, -popularity), y=popularity))+ theme(axis.text.x = element_text(angle = 45,hjust = 1)) +
xlab("Artist") +
ggtitle("Artists by total popularity since 2000")
temp <- data %>% filter(first_artist=="Taylor Swift"|first_artist=="Frank Sinatra") %>% mutate(first_artist=as.factor(first_artist))
comp_cols <- c("danceability", "energy", "loudness", "speechiness", "liveness", "instrumentalness","tempo" , "valence")
temp %>% select("first_artist", comp_cols) %>%
ggparcoord(groupColumn= 1, columns=comp_cols, order = "anyClass", alphaLines = 0.2, scale="center", splineFactor = 15)
## Note: Using an external vector in selections is ambiguous.
## ℹ Use `all_of(comp_cols)` instead of `comp_cols` to silence this message.
## ℹ See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.
pairplot <- function(data, title=NULL, columns=NULL, geom_type=geom_point, include_dist=FALSE, legend=NULL,...){
if (is.null(columns)){
columns <- 1:ncol(data)
}
col_names <- names(data)[columns]
plot_list <- list()
k <- 1
for (i in 1:length(columns)){
for (j in 1:length(columns)){
if(i==j){
p <- ggplot(data,mapping=aes_string(x=col_names[i], fill="first_artist")) + geom_histogram(data=subset(data, first_artist=="Frank Sinatra"), mapping=aes(y=..density..), colour=NA, alpha=0.5, bins=50) +
geom_histogram(data=subset(data, first_artist=="Taylor Swift"), mapping=aes(y=..density..), colour=NA, alpha=0.5, bins=50)
if(include_dist){
p <- p + geom_density(alpha=.2, fill="#FF6666")
}
plot_list[[k]] <- p
} else{
plot_list[[k]] <- ggplot(data) + geom_type(mapping=aes_string(x=col_names[i], y=col_names[j], color="first_artist"), ...)
}
k <- k+1
}
}
ggmatrix(plot_list, nrow = length(col_names), ncol = length(col_names), xAxisLabels
=col_names, yAxisLabels=col_names, title=title, legend=legend)
}
temp %>% select("first_artist", comp_cols) %>%
pairplot(columns=c(2:(length(comp_cols)+1)), geom_type=geom_point, include_dist = FALSE, size=0.5, alpha=0.3, title="Paired scatterplots for Taylor Swift and Frank Sinatra", legend=2)
detailed_comp_cols <- c("danceability", "energy", "loudness","valence")
temp %>% select("first_artist", detailed_comp_cols) %>%
pairplot(columns=c(2:(length(detailed_comp_cols)+1)), geom_type=geom_point, include_dist = FALSE, size=0.5, alpha=0.3, title="Paired scatterplots for Taylor Swift and Frank Sinatra", legend=2)
## Note: Using an external vector in selections is ambiguous.
## ℹ Use `all_of(detailed_comp_cols)` instead of `detailed_comp_cols` to silence this message.
## ℹ See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.
temp %>% group_by(first_artist) %>% summarise_if(is.numeric, median) %>% select("first_artist", detailed_comp_cols) %>% rename(group=first_artist) %>% mutate_at(vars(loudness), scales::rescale, from=c(-60, 0)) %>% as.data.frame() -> temp2
rownames(temp2) <- temp2$group
temp2 <- temp2 %>% select(-group)
temp2 <- rbind(rep(1, 4), rep(0, 4), as.data.frame(temp2))
colors_border=c( rgb(0.2,0.5,0.5,0.9), rgb(0.8,0.2,0.5,0.9))
colors_in=c( rgb(0.2,0.5,0.5,0.4), rgb(0.8,0.2,0.5,0.4) )
temp2 %>% radarchart(pcol=colors_border ,
#pfcol=colors_in ,
plwd=4 , plty=1,)
legend(x=0.7, y=1, legend = rownames(temp2[-c(1,2),]), bty = "n", pch=20 , col=colors_border , text.col = "black", cex=1.2, pt.cex=3)
comp_cols <- c("danceability", "energy", "loudness", "valence")
temp %>% select("first_artist", comp_cols) %>%
ggparcoord(groupColumn= 1, columns=comp_cols, order = "anyClass", alphaLines = 0.2, scale="center", splineFactor = 15)
p <- ggplot(data) + geom_point(aes(x=loudness, y=energy), alpha=0.5) + transition_manual(frames = year) + labs(title="{current_frame}")
animate(plot=p,
nframes=length(unique(data$year)),
fps=3
)
